适应拓扑变化的拥塞最小化网络更新策略

软件定义网络的出现推动了航空集群机载网络的发展，同时也为机载网络的一致性更新带来了挑战。针对拓扑变化导致的网络更新期间可用路径失效问题，提出适应拓扑变化的拥塞最小化网络更新策略。首先，为减少可用路径失效带来的数据包丢失，在更新之前采用重路由机制进行处理；然后，针对重路由机制可能引起的网络拥塞，设计了贪婪流迁移算法缓解拥塞；最后，通过瞬时拥塞最小化更新算法完成整个网络更新。仿真结果表明：与传统的拥塞一致性更新算法相比，所提更新策略用规则开销、瞬时网络拥塞方面的少量提高换取了网络更新期间数据包丢失的显著降低。

Congestion-minimization network update strategy for topology changes

The emergence of software-defined networking has accelerated the development of airborne networks of aviation swarm, while also presenting challenges for their consistency update. To solve the failure problem of available paths during network update caused by topology changes, a congestion-minimization network update strategy for topology changes is proposed. Firstly, to reduce the packet loss caused by the failure of available paths, the rerouting mechanism is used for processing before the update. Then, for the network congestion caused by the rerouting mechanism, the greedy flow migration algorithm is designed to reduce the network congestion. Finally, the entire network update is completed by the instantaneous congestion minimization update algorithm. The simulation results show that compared with the traditional congestion consistency update algorithm, the update strategy proposed in this paper allows a small increase in the overhead of rules and instantaneous network congestion in exchange for a significant reduction in packet loss during network update.